Exercise Reshapes Thousands of Proteins Across Blood, Saliva and Urine, New Atlas Shows

A single bout of exercise does far more than raise the heart rate and burn calories. According to a new study published in Nature Metabolism, one session of physical activity triggers thousands of coordinated protein changes that ripple through multiple body fluids at once, reshaping the molecular landscape of blood, saliva and urine in parallel. The work, led by Noam Kurgan and colleagues, presents what the authors describe as a multi-fluid exercise proteomic atlas, a resource intended to serve as a starting point for identifying the proteins that carry the health benefits of exercise from working muscles to the rest of the body.

For years, researchers have chased a deceptively simple question: why does exercise protect against so many seemingly unrelated conditions, from cardiovascular disease to metabolic disorders? The leading hypothesis centers on exerkines, the proteins and other signaling molecules released into circulation during and after physical activity. A comprehensive review by Chow and colleagues in Nature Reviews Endocrinology in 2022 framed exerkines as potential mediators of the systemic health benefits of exercise, but identifying which of the thousands of candidate molecules actually matter has remained a formidable bottleneck. The new atlas is designed to narrow that search.

The scale of the molecular response to exercise has become clear only gradually. In 2019, Kurgan and colleagues used a sequential top-down proteomic analysis to show that a single bout of high-intensity interval exercise measurably reshapes the human serum proteome. A year later, a landmark multiomic time course published in Cell by Contrepois and colleagues revealed the sheer scale and coordination of the systemic response to acute exercise, tracking molecular changes across multiple platforms in the minutes and hours following a workout. In 2023, Mi and colleagues defined the rapid kinetics of the plasma proteome after acute exercise, demonstrating that the circulating protein landscape shifts on remarkably short timescales.

Those earlier studies, however, focused primarily on blood. The new work extends the view beyond plasma and serum to saliva and urine, capturing protein dynamics in fluids that are far easier to sample repeatedly and that reflect different aspects of systemic physiology. By profiling a single bout of exercise across these fluids simultaneously, the researchers assembled an atlas of thousands of proteins whose abundance changes in a coordinated fashion, revealing that the molecular response to exercise is not confined to the bloodstream but is a genuinely multi-compartment phenomenon.

What distinguishes the new resource is not just its breadth of fluids but its integrative strategy. Measuring thousands of exercise-responsive proteins is one thing; deciding which of them are plausible regulators of health is another. To prioritize candidates, the team layered three additional types of evidence on top of the proteomic measurements. First, they integrated tissue atlases, which allow any exercise-regulated protein to be mapped back to the tissues that produce it, connecting a changing blood protein to its likely organ of origin. Second, they incorporated human genetics, asking whether genetic variation linked to disease risk also implicates the same proteins, an approach that can hint at causality rather than mere correlation. Third, they performed functional validation in vivo, testing whether prioritized candidate proteins actually produce physiological effects when manipulated experimentally.

This triangulation matters because the field of exercise biology has struggled with a signal-to-noise problem. A bout of exercise alters hydration status, hormone levels, immune cell trafficking and tissue repair processes, all of which can shift circulating protein abundances without any of those proteins being a true mediator of long-term health benefits. By requiring convergence across multiple body fluids, tissue provenance, genetic linkage and experimental validation, the atlas filters the enormous list of exercise-responsive proteins down to a set of robust candidates that the authors propose might mediate the cardiometabolic benefits of long-term exercise training.

The genetic component of the analysis is particularly consequential. Human genetics has transformed drug discovery in other fields by highlighting proteins whose naturally occurring variation is linked to disease outcomes, effectively pointing to targets that biology itself has already tested. By linking exercise-regulated proteins to genetic associations with health and disease traits, the atlas provides a way to ask which of the molecules released during exercise sit at intersections where human variation already tells us something about clinical consequence. Proteins that are both dynamically regulated by exercise and genetically tied to cardiometabolic outcomes become far stronger candidates for follow-up than proteins that merely change in abundance after a workout.

The new study arrives in the middle of a much larger coordinated effort to map the molecular consequences of exercise. The Molecular Transducers of Physical Activity Consortium, or MoTrPAC, described by Sanford and colleagues in Cell in 2020, is a large-scale initiative to chart the dynamic responses to exercise across tissues throughout the body, generating tissue-specific atlases of the molecular changes induced by training. The multi-fluid atlas complements such tissue-centered efforts by capturing the systemic, circulating dimension of the response, the molecular traffic that flows between organs during and after activity. Together, these resources sketch a body-wide picture in which exercise is understood not as a single signal but as a coordinated, multi-tissue conversation conducted partly through the blood, saliva and urine.

The clinical implications are substantial. If the exercise-regulated proteins prioritized in the atlas can be validated as genuine mediators of cardiometabolic health, they could become targets for therapies that mimic the benefits of physical activity in people who cannot exercise, whether because of disability, chronic illness or advanced age. They could also serve as biomarkers, objective molecular readouts of how an individual responds to a workout, potentially enabling exercise prescriptions that are monitored and tuned at the molecular level. The use of saliva and urine alongside blood is especially relevant here, since less invasive sampling makes repeated molecular monitoring far more practical in real-world settings.

The authors are careful to frame the atlas as a starting point rather than a final answer. Prioritizing candidate exerkines is the first step in a longer pipeline of mechanistic work, functional studies and eventual clinical testing. But by bringing together multi-fluid proteomics, tissue atlases, human genetics and in vivo validation in a single resource, the study offers the exercise biology community something it has lacked: a systematically ranked shortlist of molecular suspects in one of medicine’s most enduring mysteries, the question of how a workout communicates its benefits to virtually every organ in the body. As the field moves from cataloging the molecular choreography of exercise to manipulating it, atlases of this kind are likely to become the maps that guide the way.

Subject of Research: Multi-fluid proteomic mapping of the acute exercise response and prioritization of exercise-regulated proteins linked to health and disease

Article Title: A multi-fluid exercise proteomic atlas reveals regulators of health and disease

Article References: A multi-fluid exercise proteomic atlas reveals regulators of health and disease. (n.d.). https://doi.org/10.1038/s42255-026-01642-5

Image Credits: AI Generated

DOI: 10.1038/s42255-026-01642-5

Keywords: exercise, proteomics, exerkines, blood, saliva, urine, Nature Metabolism, cardiometabolic health, human genetics, tissue atlases, MoTrPAC, molecular transducers